Related Experiment Videos
A neural network model of the vestibulo-ocular reflex using a local synaptic learning rule
1Department of Biomedical Engineering, Johns Hopkins University, School of Medicine, Baltimore, Maryland 21287-9131.
Summary
Vertebrates maintain vision during head movements using the vestibulo-ocular reflex. This study models the neural network in the brainstem that integrates eye-velocity signals into eye-position commands for clearer sight.
Area of Science:
- Neuroscience
- Ophthalmology
- Computational Biology
Background:
- The vestibulo-ocular reflex (VOR) is crucial for maintaining visual stability during head motion in vertebrates.
- The VOR integrates semicircular canal input (eye velocity) to generate extraocular muscle commands (eye position).
- This neural integration occurs within a specific network in the caudal pons.
Purpose of the Study:
- To propose a computational model of the neural network responsible for VOR.
- To investigate the role of positive feedback and lateral inhibition in this network.
- To adapt the model into a learning network with a physiological synaptic learning rule.
Main Methods:
- Development of a computational model for the VOR neural network.
- Incorporation of positive feedback and lateral inhibition mechanisms.
- Adaptation of the model to a synaptic learning rule using local information.
Main Results:
- A model simulating VOR neural integration was developed.
- The model incorporates positive feedback and lateral inhibition.
- A novel, physiologically plausible synaptic learning rule was created for the model.
Conclusions:
- The proposed model effectively simulates the neural integration required for the VOR.
- Positive feedback and lateral inhibition are key components of this neural circuit.
- The developed learning rule enhances the model's physiological relevance for understanding VOR adaptation.